Structure and dynamics of a molecular hydrogel derived from a tripodal cholamide

Article abstract of DOI:10.1021/ja046788t

Tripodal cholamide 1 is a supergelator of aqueous fluids. A variety of physical techniques, including cryo-transmission electron microscopy (TEM), circular dichroism (CD), steady-state fluorescence, time-resolved fluorescence, and dynamic light-scattering, were employed to understand the structure and dynamics of the gel. Fluorescent probes [ANS (8-anilinonaphthalene-1-sulfonic acid) and pyrene] reported two critical aggregation concentrations (CAC ₁ and CAC₂) of 1 in predominantly aqueous media, with the minimum gel concentration (MGC) being close to CAC₂. Fluorescence lifetime measurements with pyrene revealed ineffective quenching of pyrene fluorescence by oxygen, possibly caused by slower Brownian diffusion due to the enhanced viscosity in the gel phase. The study of the gelation kinetics by monitoring the ultrafast dynamics of ANS revealed a progressive increase in the aggregate size and the microviscosity of the aqueous pool encompassed by the self-assembled fibrillar network (SAFIN) during the gelation. The striking difference between microviscosity and bulk (macroscopic) viscosity of the gel is also discussed.

Full text of DOI:10.1021/ja046788t

Published on Web 11/10/2004
Structure and Dynamics of a Molecular Hydrogel Derived from
a Tripodal Cholamide
†
,†
‡
‡
Samrat Mukhopadhyay, Uday Maitra,* Ira, Guruswamy Krishnamoorthy,
¶
¶
Judith Schmidt, and Yeshayahu Talmon
Contribution from the Department of Organic Chemistry, Indian Institute of Science,
Bangalore 560 012 India, Department of Chemical Sciences, Tata Institute of Fundamental
Research, Mumbai 400 005, India, and Department of Chemical Engineering, Technion-Israel
Institute of Technology, Haifa 32000, Israel
Received June 1, 2004; E-mail: maitra@orgchem.iisc.ernet.in
Abstract: Tripodal cholamide 1 is a supergelator of aqueous fluids. A variety of physical techniques, including
cryo-transmission electron microscopy (TEM), circular dichroism (CD), steady-state fluorescence, time-
resolved fluorescence, and dynamic light-scattering, were employed to understand the structure and
dynamics of the gel. Fluorescent probes [ANS (8-anilinonaphthalene-1-sulfonic acid) and pyrene] reported
two critical aggregation concentrations (CAC
1
2
and CAC ) of 1 in predominantly aqueous media, with the
minimum gel concentration (MGC) being close to CAC
2
. Fluorescence lifetime measurements with pyrene
revealed ineffective quenching of pyrene fluorescence by oxygen, possibly caused by slower Brownian
diffusion due to the enhanced viscosity in the gel phase. The study of the gelation kinetics by monitoring
the ultrafast dynamics of ANS revealed a progressive increase in the aggregate size and the microviscosity
of the aqueous pool encompassed by the self-assembled fibrillar network (SAFIN) during the gelation. The
striking difference between microviscosity and bulk (macroscopic) viscosity of the gel is also discussed.
tides,6 carbohydrate derivatives, gemini surfactants/bola-
7
1
. Introduction
8
9
10
amphiphiles, bile acids, and others have been shown to act
as gelators of aqueous fluids. These hydrogels are of great
The formation of gels in water has been well documented
1
with polymers and biopolymers. Such materials are known as
hydrogels or aqueous gels. Gelation with this class of molecules
is believed to occur by chemical and/or physical cross-linking
of polymeric chains, leading to the formation of a highly
intertwined three-dimensional network, which restrains water
molecules by surface tensional forces. When the driving forces
for gelation involve molecular self-assembly through noncova-
lent physical forces, the resulting hydrogels are termed as
(5) (a) Imae, T.; Takahashi, Y.; Muramatsu, H. J. Am. Chem. Soc. 1992, 114,
3414. (b) Fuhrhop, J.-H.; Spiroski, D.; Boettcher, C. J. Am. Chem. Soc.
1993, 115, 1600. (c) Menger, F. M.; Caran, K. L. J. Am. Chem. Soc. 2000,
122, 11679. (d) Estroff, L. A.; Hamilton, A. D. Angew. Chem., Int. Ed.
2000, 39, 3447. (e) Frkanec, L.; Jokic, M.; Makarevic, J.; Wolsperger, K.;
Zˇ inic, M. J. Am. Chem. Soc. 2002, 124, 9716. (f) Suzuki, M.; Yumoto,
M.; Kimura, M.; Shirai, H.; Hanabusa, K. Chem. Commun. 2002, 884. (g)
Heeres, A.; van der Pol, C.; Stuart, M.; Friggeri, A.; Feringa, B. L.; van
Esch, J. J. Am. Chem. Soc. 2003, 125, 14252. (h) Suzuki, M.; Yumoto,
M.; Kimura, M.; Shirai, H.; Hanabusa, K. Chem.sEur. J. 2003, 9, 348. (i)
van Bommel, K. J. C.; van der Pol, C.; Muizebelt, I.; Friggeri, A.; Heeres,
A.; Meetsma, A.; Feringa, B. L.; van Esch, J. Angew. Chem., Int. Ed. 2004,
2
physical gels or molecular gels. The past decade has witnessed
43, 1663.
a rapid growth in this area of nonpolymeric gels. Compared to
(
6) (a) Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Proc. Natl. Acad. Sci. U.S.A.
2002, 99, 5133. (b) Aggeli, A.; Bell, M.; Boden, N.; Keen, J. N.; Knowles,
P. F.; McLeish, T. C. B.; Pitkeathly, M.; Radford, S. E. Nature 1997, 386,
259. (c) Collier, J. H.; Hu, B.-H.; Ruberti, J. W.; Zhang, J.; Shum, P.;
Thompson, D. H.; Messersmith, P. B. J. Am. Chem. Soc. 2001, 123, 9463.
3
organogels, reports on molecular hydrogels have been scarce
4
but have been rapidly increasing in recent years. A wide variety
5
of molecular species involving amino acid derivatives, polypep-
(
d) Xing, B.; Yu, C.-W.; Chow, K.-H.; Ho, P.-L.; Fu, D.; Xu, B. J. Am.
Chem. Soc. 2002, 124, 14846. (e) Schneider, J. P.; Pochan, D. J.; Ozbas,
B.; Rajagopal, K.; Pakstis, L.; Kretsinger, J. J. Am. Chem. Soc. 2002, 124,
†
Indian Institute of Science.
Tata Institute of Fundamental Research.
Technion-Israel Institute of Technology.
1
5030. (f) Pochan, D. J.; Schneider, J. P.; Kretsinger, J.; Ozbas, B.;
‡
Rajagopal, K.; Haines, L. J. Am. Chem. Soc. 2003, 125, 11802. (g) Claussen,
R. C.; Rabatic, B. M.; Stupp, S. I. J. Am. Chem. Soc. 2003, 125, 12680.
¶
(
1) (a) Terech, P. Encyclopedia of Surface and Colloid Science; Marcel
Dekker: New York, 2002; pp 2299-2319. (b) Guenet J. M. Thermor-
eVersible Gelation of Polymers and Biopolymers; Academic Press: New
York, 1992.
(7) (a) Pfannemuller, B.; Welte, W. Chem. Phys. Lipids 1985, 37, 227. (b)
Furhorp, J.-H.; Schnieder, P.; Rosenberg, J.; Boekema, E. J. Am. Chem.
Soc. 1987, 109, 3387. (c) Bhattacharya, S.; Acharya, S. N. G. Chem. Mater.
1999, 11, 3504. (d) Kobayashi, H.; Friggeri, A.; Koumoto, K.; Amaike,
M.; Shinkai, S.; Reinhoudt, D. N. Org. Lett. 2002, 4, 1423. (e) Jung, J. H.;
Shinkai, S.; Shimizu, T. Chem.sEur. J. 2002, 8, 2684. (f) Kiyonaka, S.;
Shinkai, S.; Hamachi, I. Chem.sEur. J. 2003, 9, 976.
(8) (a) Newkome, G. R.; Baker, G. R.; Arai, S.; Saunders, M. J.; Russo, P. S.;
Theriot, K. J.; Moorefield, C. N.; Rogers, L. E.; Miller, J. E.; Lieux, T. R.;
Murray, M. E.; Phillips, B.; Pascal, L. J. Am. Chem. Soc. 1990, 112, 8458.
(b) Oda, R.; Huc, I.; Candau, S. J. Angew. Chem., Int. Ed. 1998, 37, 2689.
(c) Iwaura, R.; Yoshida, K.; Masuda, M.; Yase, K.; Shimizu, T. Chem.
Mater. 2002, 14, 3047. (d) Menger, F. M.; Peresypkin, A. J. Am. Chem.
Soc. 2003, 125, 5340.
(
2) Molecular Gels; Terech, P., Wiess, R. G., Eds.; Kluwer Academic
Publishers: The Netherlands, 2004.
(
3) For recent reviews on organogels, see: (a) Terech, P.; Weiss, R. G. Chem.
ReV. 1997, 97, 3133. (b) van Esch, J. H.; Feringa, B. L. Angew. Chem.,
Int. Ed. 2000, 39, 2263. (c) Abdallah, D. J.; Weiss, R. G. AdV. Mater.
2
000, 12, 1237. (d) See ref 2.
(
4) (a) Bhattacharya, S.; Maitra, U.; Mukhopadhyay, S.; Srivastava, A. In
Molecular Gels; Terech, P., Wiess, R. G., Eds.; Kluwer Academic
Publishers: The Netherlands, 2004. (b) Estroff, L. A.; Hamilton, A. D.
Chem. ReV. 2004, 104, 1201.
10.1021/ja046788t CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 15905-15914
9
15905

A R T I C L E S
Chart 1
Mukhopadhyay et al.
plication of such hydrogels in the template-directed preparation
of nanotubes of inorganic oxides and sulfates has also been
1
6
reported. Herein, we disclose the synthetic protocols and
detailed studies on the aggregation behavior of tripodal chol-
amide 1, deoxycholamide 2, and their monomeric analogues, 3
and 4 (Chart 1). A combination of microscopic and spectro-
scopic techniques was used to understand the structure and
dynamics of gels. Also, we provide insights into the time course
of the gelation process using a double-kinetic experiment.
2. Results
2
.1. Synthesis. The tripodal bile acid derivatives (nonahy-
droxy 1 and hexahydroxy 2) were prepared in three steps from
cholic and deoxycholic acid, respectively (Scheme 1). Formyl-
Scheme 1 a
importance due to the fact that they are potential materials for
biomedical applications, such as drug-delivery systems, tissue
engineering, semi-wet biomaterials for protein microarray, and
1
1
so forth. Recently, the advantages of using the molecular
hydrogels over traditional polymeric hydrogels have been
reviewed.12 In addition to potential applications, gels are
materials with intriguing features owing to the coexistence of
solid (the networked fibrous structure) and liquid (entrapped
solvent molecules) phases. These gels are viscoelastic materials
since they display properties of both solids (elasticity) and
13
liquids (viscosity). Structural analysis of these materials is not
straightforward since they do not lend themselves to studies at
atomic resolutions. Applications of a variety of physical
techniques are required to gain insights into the complex nature
2
-4
a
of the gel phase.
(a) HCO2H, rt, 5 h. (b) DCC-DMAP/N[CH2CH2NH2]3 in DCM, 12
h. (c) DCC-DMAP/Me2NCH2CH2NH2 in DCM, 5 h. (d) With 5% KOH-
We have recently described the efficient gelation ability of a
novel tripodal cholamide in aqueous media.14 The gel formation
was associated with the formation of hydrophobic pockets in a
predominantly aqueous medium. The rotational dynamics of
polarity-sensitive organic dyes partitioned in the hydrophobic
region and in the aqueous phase of the gel was determined using
MeOH.
protected bile acids were coupled with tren to obtain the tripodal
architecture. In the next step, the formyl groups were cleaved
to obtain the free hydroxy tripodal bile acid derivatives. N,N-
Dimethyl(ethylenediamine) was used to obtain monomeric
analogues.
1
5
the picosecond time-resolved fluorescence method. The ap-
2
.2. Behavior in Aqueous Media. Nonahydroxy 1 and
(
9) (a) Schryver, S. B. R. Soc. Proc., Ser. B 1914, 87, 366. (b) Schryver, S. B.
R. Soc. Proc., Ser. B 1916, 89, 361. (c) Sobotka, H.; Czeczowiczka, N. J.
Colloid Sci. 1958, 13, 188. (d) Rich, A.; Blow, D. M. Nature 1958, 182,
hexahydroxy 2 were insoluble in water. The protonated salts,
such as 1‚HCl and 2‚HCl, have limited solubility in water.
However, upon the addition of a small amount (5-20%) of
organic cosolvents (e.g., EtOH, MeOH, CH3CN, DMSO, DMF,
and acetone), clear solutions were obtained. Compound 1 formed
a transparent gel (Figure 1A) from such solutions in a few
minutes, but 2 failed to form a gel under all of the conditions
we studied. Monomeric analogues (3 and 4) also did not act as
gelators of aqueous fluids. They formed either precipitates or
clear solutions depending on the concentration of the compound
and the composition of the cosolvents. The “best” gels
4
23. (e) Blow, D. M.; Rich, A. J. Am. Chem. Soc. 1960, 82, 3566. (f)
Igimi, H.; Carey, M. J. Lipid Res. 1980, 21, 72. (g) Terech, P.; Smith, W.
G.; Weiss, R. G. J. Chem. Soc., Faraday Trans. 1996, 92, 3157. (h) Jover,
A.; Meijide, F.; N u´ n˜ ez, E. R.; Tato, J. V. Langmuir 1996, 12, 1789. (i)
Lopez, F.; Samseth, J.; Mortensen, K.; Rosenqvist, E.; Rouch, J. Langmuir
1
996, 12, 6188.
(
10) (a) Fuhrhop, J.-H.; Demoulin, C.; Rosenberg, J.; Boettcher, C. J. Am. Chem.
Soc. 1990, 112, 2827. (b) Haines, S. R.; Harrison, R. G. Chem. Commun.
2
002, 2846. (c) Marmillon, C.; Gauffre, F.; Gulik-Krzywicki, T.; Loup,
C.; Caminade, A.-M.; Majoral, J.-P.; Vors, J.-P.; Rump, E. Angew. Chem.,
Int. Ed. 2001, 40, 2626. (d) Park, S. M.; Lee, Y. S.; Kim, B. H. Chem.
Commun. 2003, 2912.
(
11) (a) Lee, K. Y.; Moony, D. J. Chem. ReV. 2001, 101, 1869. (b) Miyata, T.;
Uragami, T.; Nakamae, K. AdV. Drug DeliVery ReV. 2002, 54, 79. (c)
Kiyonaka, S.; Sada, K.; Yoshimura, I.; Shinkai, S.; Kato, N.; Hamachi, I.
Nat. Mater. 2004, 3, 58.
(
transparent and thermally stable) were obtained in acetic acid-
(
12) Tiller, J. C. Angew Chem., Int. Ed. 2003, 42, 3072.
13) Ross-Murphy, S. B. Physical Techniques for the Study of Food Biopolymers;
Blackie: London, 1994.
(
(15) Mukhopadhyay, S.; Ira; Krishnamoorthy, G.; Maitra, U. J. Phys. Chem. B.
2003, 107, 2189.
(16) Gundiah, G.; Mukhopadhyay, S.; Tumkurkar, U. G.; Govindaraj, A.; Maitra,
U.; Rao, C. N. R. J. Mater. Chem. 2003, 13, 2118.
(
14) Maitra, U.; Mukhopadhyay, S.; Sarkar, A.; Rao, P.; Indi, S. S. Angew.
Chem., Int. Ed. 2001, 40, 2281.
15906 J. AM. CHEM. SOC.
9
VOL. 126, NO. 48, 2004

Molecular Hydrogel from a Tripodal Cholamide
A R T I C L E S
Figure 1. (A) Transparent gel: [1] ) 9 mM in 20% AcOH-H2O. (B)
Luminescent gel: [1] ) 5 mM and [ANS] ) 30 µM.
Figure 3. CD spectra of the gel at 25 °C (a) and of the sol at 75 °C (b)
from 1‚HCl (3.0 mM in 20% EtOH-water). The inset shows the variable
temperature CD of the gel (heating and cooling cycle).
Figure 2. Cryo-TEM images of the gel (0.75 mM) in 20% AcOH-water
at two different magnifications. Bars represent 100 nm (a) and 50 nm (b).
Table 1. Minimum Gel Concentrations (MGC) of 1 as a Function
of Solvent Composition at Room Temperature
solvent composition (v/v)
MGC (mM)
2
0% AcOH-water
% ACOH-water
.05% AcOH-water
.01% AcOH-water
0.75
0.60
0.30
0.15
1
0
0
Figure 4. Change in the fluorescence intensity of ANS (10 µM) and the
I3/I1 ratio (pyrene fluorescence, [pyrene] ) 0.5 µM) as a function of the
gelator concentration at 25 °C.
water systems ranging from 0.01 to 30% AcOH in water,
depending on the gelator concentration (Table 1). These gels
were thermoreversible and thixotropic in nature, and they were
stable for more than 3 years when kept in sealed tubes. Gels
were formed under remarkably low concentrations of the gelator.
from the hydrochloride salt ([1‚HCl] ) 3 mM in 20% EtOH-
water). A negative CD band (amide chromophore) was observed
for both isotropic solution ([1‚HCl] ) 3 mM in neat EtOH)
and gel. The CD band intensity at 216 nm was twice as much
(21 mdeg) for gel than the intensity for the isotropic solution
of 1‚HCl in neat EtOH (10 mdeg), which does not form a gel
(not shown). Figure 3 shows that the intensity of the CD band
diminishes to about 11 mdeg at the gel melting temperature
(Tgel ≈ 55 °C). The variable temperature CD showed hysteresis
(nonidentical heating and cooling curves).
1
7
The minimum gel concentration was as low as 0.15 mM,
implying the immobilization (apparent rigidification) of more
5
than 3 × 10 water molecules by one molecule of 1. To the
best of our knowledge, among the hydrogelators known to date,
1
forms gels at the lowest gelator concentration.
.3. Electron Microscopy. Cryogenic temperature transmis-
2
sion electron microscopy (cryo-TEM) images of vitrified
specimens of the gel formed in a solution of 0.75 mM 1 in
2
.5. Steady-State Fluorescence of Pyrene. Pyrene was used
as an external fluorescent probe to determine the critical
aggregation concentration. The vibronic fine structure of the
pyrene fluorescence is indicative of local polarity of the binding
site. The plot (Figure 4) of the ratio of two vibronic bands (I3/
I1) versus the gelator concentration (in 20% AcOH-water)
suggested two critical aggregation concentration (CAC) ranges
2
0% AcOH-water are shown in Figure 2. At the two magni-
fications shown, the gel appears as a well-developed intertwined
network. The higher magnification view of Figure 2b reveals
that the network is made of very thin, flat ribbons (2-5 nm
wide). That those are indeed flat ribbons can be deduced from
the uniformity of optical density across each one of them. One
should realize that in the TEM, all structures in the thin
specimen, approximately 100 nm thick, are projected and seen
clearly in the image. This gives the impression of the presence
of much more polymer in the gel than its actual concentration.
(
CAC1 ≈ 0.4 mM and CAC2 ≈ 0.8 mM). The gelation was
visually observed at 0.75 mM. No excimer formation was
observed in the gel phase. With lower amounts of AcOH (e.g.,
1
(
% AcOH-water), the minimum gel concentration was lower
0.6 mM) and I3/I1 values were higher (Table 2). Another
interesting observation was the formation of an excimer band
2
.4. Circular Dichroism. To investigate the chiral structure
of gels, CD experiments were performed on the gel derived
(
centered on 480 nm) in the pyrene fluorescence spectrum for
(
micellar) aggregates of 1 in 1% AcOH-water below a
(
17) Partial gelations could still be observed at micromolar concentrations (75
µM) of 1, but these gels were unstable at room temperature.
minimum gel concentration (i.e., below 0.6 mM). In the gel
J. AM. CHEM. SOC.
9
VOL. 126, NO. 48, 2004 15907

A R T I C L E S
Mukhopadhyay et al.
Table 2. Fluorescence Lifetimes and I3/I1 Values for Pyrene in
Solutions, in Aggregates, and in Gels at 25 °C
sample
(ns)a
b
(
[pyrene]
≈
0.5
µM)
τ
3 1
I /I
water
0% AcOH-water
135
137
170
0.58
0.61
0.75
2
aggregate of 1 (0.6 mM)
in 20% AcOH-water
gel from 1 (5.3 mM)
in 20% AcOH-water
aggregate of 1 (42 µM)
in 1% AcOH-water
gel from 1 (0.8 mM)
in 1% AcOH-water
226
1.10
1.42
1.53
239 (68)c
324
Figure 6. Temperature dependence of the steady-state fluorescence
anisotropy (rss) of ANS in the gel from nonahydroxy salt (3.0 mM 1‚HCl
in 20% EtOH-water).
a
Errors are <(5 ns. b Errors are <(0.05. c Excimer lifetime.
Figure 7. Change in the average translational diffusion coefficient (Dt) of
the aggregates as a function of time during the gelation at 25 °C. [1] ) 5.3
mM. Bars represent the measured polydispersity.
and in the gel, respectively), which are close to those observed
in deoxygenated media (∼400 ns), indicating ineffective
quenching by oxygen. Excimer lifetime was found to be ∼68
ns in the micellar aggregate of 1 in 1% AcOH-water.
Figure 5. Fluorescence spectra of pyrene in (a) gel ([1] ) 1.7 mM) and
(
b) micellar aggregate ([1] ) 25 µM) in 1% AcOH-water. I1, I3, and E are
the first vibronic band, the third vibronic band, and the excimer band,
respectively. The inset shows the variation of the excimer intensity (E at
4
75 nm) as a function of [1] below MGC.
2
.8. Dynamic Light-Scattering. The average translational
diffusion coefficient (Dt) of the aggregates is inversely propor-
tional to the aggregate size and the viscosity by the Stokes-
Einstein equation for translational diffusion (eq 1).
phase (i.e., at a higher concentration of 1), this band disappeared
Figure 5).
.6. Steady-State Fluorescence of ANS. Using steady-state
fluorescence titration, we have studied the interaction of ANS
8-anilino-1-naphthalene sulfonic acid) with increasing amounts
of 1 to monitor the onset of the aggregation process. The plot
Figure 4) of the intensity of ANS fluorescence versus gelator
concentration (in 20% AcOH-water) also suggested two critical
aggregation concentration ranges (∼0.3 and ∼0.7 mM). Steady-
state fluorescence anisotropy of ANS in gel decreases as a
function of temperature and remains almost constant beyond
the gel melting temperature. This indicates higher mobility of
ANS in the sol phase compared to that in the gel phase. The
variable temperature anisotropy (Figure 6) exhibits hysteresis
in the sol-to-gel phase transition, as observed in the variable
temperature CD.
(
2
kT
6πηRh
D )
(1)
t
(
(
where η is the solvent viscosity, and Rh is the mean hydrody-
namic radius of the diffusing unit.
For micellar aggregates (0.6 mM 1 in 20% AcOH-water),
-9
2
-1
Dt was about 70 × 10 cm s . The micellization was fast
and did not show any change in Dt with time. On the contrary,
the gelation (5.3 mM 1 in 20% AcOH-water) process was
slower and could be monitored by the dynamic light-scattering
-9
technique. Dt was found to decrease (from 38 × 10 to 14 ×
-9
2 -1
1
0
cm s ) as a function of time during the gelation process
(Figure 7). No significant change was observed after 2 h. Also,
2
.7. Time-Resolved Fluorescence. Fluorescence lifetime (τ)
the polydispersity, which is related to the heterogeneity in the
aggregate structure in terms of the aggregation number, was
found to decrease as a function of time during gelation. The
data clearly show the progressive increase in the homogeneity
of the aggregate structure during the course of gelation.
2.9. Time-Resolved Fluorescence Anisotropy of ANS
during Gelation. The time-resolved decay of the fluorescence
anisotropy [r(t)] of ANS, in both water and 20% AcOH-water,
of pyrene in nondeoxygenated water was found to be ∼135 ns.
In the presence of increasing amounts (1-20% v/v) of acetic
acid in water, change in the fluorescence lifetime (τ) of pyrene
was insignificant (135-137 ns). However, τ increased in the
aggregates and gels derived from 1 in 20% AcOH-water (Table
2
). In 1% AcOH-water, even longer fluorescence lifetimes were
observed for pyrene (239 and 324 ns in the micellar aggregate
15908 J. AM. CHEM. SOC.
9
VOL. 126, NO. 48, 2004

Molecular Hydrogel from a Tripodal Cholamide
A R T I C L E S
Table 3. Recovered Rotational Correlation Times of ANS (φb for
bound and φf for free) from the Time-Resolved Anisotropy
Experiments in Isotropic Solution, Premicellar Aggregate, Micellara
Aggregate, and Gel from 1 under Equilibrium Conditions at 25 °C
Table 4. Change in the Rotational Correlation Times of ANS as a
Function of Time during the Gelation (at 25 °C) Obtained from the
Double-Kinetics Experiments
Rotational Correlation Timesa
time after
sample
[1] (mM)
φ
b
(ns)
φ
f
(ns)
b
mixing (min)
φ
b
(ns)
φ
f
(ns)
η
aq (cP)
2
0% AcOH-H2O
0.1
0.1
4
3.7
3.9
4.8
7.4
11.0
12.1
0.2
0.4
0.6
0.7
1.0
1.0
2.7
5.3
8.0
9.3
13.3
13.3
below CAC1 of 1
in 20% AcOH-H2O
above CAC1 and
below CAC2
above CAC2 and
above MGC
0.2
0.6
5.3
2.1
4.9
1
2
5
0
3
5
26
60
0.2
1.0
1
1
12.1
a
Errors in the analysis are <(1.0 ns (for φb) and <(0.1 ns (for φf).
ηaq is the microviscosity felt by ANS.
a
b
Errors in the analysis were <10%.
the gelation process was estimated from the Stokes-Einstein
equation for rotational diffusion (eq 2) using the rotational
correlation time of the free probe (φf).
yielded a single-exponential decay with a rotational correlation
time of about 0.1 ns. The molecular volume of ANS was
calculated (ab initio) and found to be 282.9 Å . The calculated
3
rotational correlation time of ANS in 20% AcOH-water
ηV
kT
(viscosity ) 1.2 cP at 25 °C) was found to be 84 ps, obtained
φ )
(2)
by the Stokes-Einstein equation for rotational diffusion (eq 2),
which is close to the experimentally determined value, 90 (
where V is the volume of the rotating unit, and η is the
microviscosity experienced by ANS.
1
0 ps (∼0.1 ns). When data were collected for compound 1,
dissolved in 20% AcOH-water, [r(t)] of ANS was more
complex and could be satisfactorily explained by a bi-
exponential decay function. Since this behavior could be
indicative of two populations of the fluorescent probe, [r (t)]
was analyzed by a two-population model (see section 5.8 for
the method). The intensity decay analysis (fluorescence life-
times) also suggested the presence of two populations; thus,
the two recovered rotational correlation time data could be
assigned to free (exposed to aqueous phase) and bound (to
hydrophobic pockets) forms of ANS. This is also justified by
the short and long values of the rotational correlation times.
Similar data were recovered when the anisotropy decay curves
were analyzed by a single-population model (as a sum of two
exponentials).
3
. Discussion
3.1. Nanoscale Structure of Gel. Cryo-transmission electron
microscopy was used to visualize the nanoscale structure of the
gel. The advantages of using the cryo-TEM are 2-fold. The fast
freezing of the samples upholds the hydrated morphology, and
the staining agents are not required because of the intrinsic
29
contrast between vitrified ice and the organic material. A TEM
image of the gel shows a network structure composed of
entangled nanofibers (Figure 2) forming meshes, which are
responsible for immobilizing the solvent molecules. These fibers
are formed by one-dimensional growth of the supramolecular
assembly of 1 in aqueous acids.
3
.2. Amplification of Chirality in the Gel State. Circular
At a concentration of 0.2 mM 1 (below the first critical
aggregation concentration, CAC1), ANS has a longer rotational
correlation time of 2.1 ns, which increased to 4.9 ns with the
increase in the concentration of 1 to 0.6 mM (above CAC1).
The shorter component (e0.2 ns for the aqueous population of
ANS) did not show a significant change (Table 3). Further
increase in the gelator concentration to 5.3 mM (well above
MGC) significantly changed the dynamics of both bound (12.1
ns) and free (1.0 ns) ANS. Therefore, we undertook the “double-
kinetic” experiments (fluorescence anisotropy decay kinetics as
a function of gelation kinetics) as follows. Compound 1 was
dissolved in AcOH; an appropriate amount of water was added,
and the time evolutions of both φb and φf were determined
dichroism spectra indicated that molecules are self-assembled
in an asymmetric manner (e.g., helical) in the gel phase causing
the amplification of the chirality in the gel phase, which can be
attributed to the supramolecular chirality in the gel state. Around
the melting temperature of the gel (∼55 °C), the sharp decrease
in the ellipticity indicates the loss of (supramolecular) chirality
upon gel melting (Figure 3). The CD (at 216 nm) of the sol (11
(
21) (a) Nakajima, A. Bull. Chem. Soc. Jpn. 1973, 46, 2602. (b) Hara, K.; Ware,
W. Chem. Phys. 1981, 51, 61. (c) Karpovich, D. S.; Blanchard, G. J. J.
Phys. Chem. 1995, 99, 3951.
(
22) Slavik, J. Biochim. Biophys. Acta 1982, 694, 1.
(
23) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Plenum Press:
New York, 1999.
(
24) (a) Maitra, U.; Kumar, P. V.; Chandra, N.; D’Souza, L. J.; Prasann, M. D.;
Raju, A. R. Chem. Commun. 1999, 595. (b) Maitra, U.; Kumar, P. V.;
Sangeetha, N. M.; Babu, P.; Raju, A. R. Tetrahedron: Asymmetry 2001,
12, 477. (c) Babu, P.; Sangeetha, N. M.; Kumar, P. V.; Maitra, U.; Rissanen,
K.; Raju, A. R. Chem.sEur. J. 2003, 9, 1922.
(
Table 4). The investigation on the gelation kinetics using time-
25
dependent anisotropy (a double-kinetics approach) indicated
a gradual change in both the rotational correlation times (φb
for “bound” ANS and φf for “free” ANS). The change in the
microviscosity of the aqueous phase (from ∼3 to 13 cP) during
(25) (a) Saxena, A.; Udgaonkar, J. B.; Krishnamoorthy, G. In Applications of
Fluorescence Spectroscopy; Hof, M., Ed.; Springer-Verlag: New York,
2004. (b) Krishnamoorthy, G. Indian J. Biochem. Biophys. 2003, 40, 147.
(c) Maiti, N. C.; Krishna, M. M. G.; Britto, P. J.; Periasamy, N. J. Phys.
Chem. B 1997, 101, 11051. (d) Lakshmikanth, G. S.; Krishnamoorthy, G.
Biophys. J. 1999, 77, 1100.
(
18) (a) Kalyanasundaram, K. In Photochemistry in Organized & Constrained
(26) Rotational relaxation time of ANS was estimated to be 161 ns (using the
Perrin equation) in pure glycerol at 22.6 °C: Treloar, F. E. Chem. Scr.
1976, 10, 215.
Media; Ramamurthy, V., Ed.; VCH Publishers: New York, 1991; Chapter
2
. (b) Bohne, C.; Redmond, R. W.; Scaiano, J. C. In Photochemistry in
Organized & Constrained Media; Ramamurthy, V., Ed.; VCH Publishers:
New York, 1991; Chapter 3. (c) Kalyansundaram, K.; Thomas, J. K. J.
Am. Chem. Soc. 1977, 99, 2039. (d) Gouin, S.; Zhu, X. X. Langmuir 1998,
(27) Cortese, F.; Bauman, L. J. Am. Chem. Soc. 1935, 57, 1393.
(28) Takahashi, A.; Sakai, M.; Kato, T. Polym. J. 1980, 12, 335.
(29) (a) Talmon, Y. Cryogenic Temperature Transmission Electron Microscopy
in the Study of Surfactant Systems. In Modern Characterization Methods
of Surfactant Systems; Binks, B. P., Ed.; Marcel Dekker: New York, 1999;
Chapter 5, pp 147-178. (b) Estroff, L. A.; Leiserowitz, L.; Addadi, L.;
Weiner, S.; Hamilton, A. D. AdV. Mater. 2003, 15, 38.
1
4, 4025.
(
(
19) Sahoo, L.; Sarangi, J.; Misra, P. K. Bull. Chem. Soc. Jpn. 2002, 75, 859.
20) Kweon, M.; Lee, Y. H.; Ahn, B. T.; Lee, M. Bull. Korean Chem. Soc.
1
996, 17, 158.
J. AM. CHEM. SOC.
9
VOL. 126, NO. 48, 2004 15909

A R T I C L E S
Scheme 2
Mukhopadhyay et al.
(bimolecular quenching rate constant, kq) the long-lived S1
(eq 3).
1
τ )
(3)
k + k + k [O ]
r
nr
q
2
The time-resolved decay of the fluorescence intensity of
pyrene provides more insights into the microenvironment of the
aggregates. The fluorescence lifetime of pyrene (τ) increases
when pyrene is solubilized in the micellar aggregate or in the
gel phase (Table 2). This increase in τ is attributed to the
decrease in both of the rate constants (knr and kq). It is expected
that quenching becomes inefficient since the diffusion coefficient
of oxygen (D, which is inversely proportional to viscosity, η)
is lowered due to the increase in the viscosity in the gel phase
mdeg, which remains constant beyond Tgel) is close to that
observed for an isotropic solution (∼10 mdeg) of 1 in neat
EtOH, suggesting nearly complete disruption of the aggregation.
3.3. Local Polarity and the Critical Aggregation Concen-
trations. The analysis of emission spectra from fluorescent
probes has provided important details on the structure, dynamics,
and transport in organized assemblies.18 A hydrophobic guest
molecule, such as pyrene, is known to bind to the hydrophobic
(eq 5). Our results indeed indicate that the pyrene is more
effectively protected from the quenching by oxygen in the gel
phase. Thus, in the gel phase, the major modification occurs
in the quenching term (kq[O2] component) of eq 3. The root-
1
8
core of a micellar aggregate, including bile salt micelles.
Fluorescence spectra of such fluorescent guest molecules carry
information on the nature of the hydrophobic interior. The ratio
of two vibronic bands (I3/I1) in the fluorescence spectrum of
pyrene is indicative of the polarity (dielectric constant) of the
binding site in the micelles/aggregates.18c The critical aggrega-
tion concentrations (CACs) of 1 were determined from the
dependence of I3/I1 on the gelator concentration from the
equilibrium experiments. Two inflections (Figure 4) correspond
to at least two steps of aggregation (primary and secondary
aggregation). From the correlation of I3/I1 of pyrene and the
ET(30) values reported by others, we infer that the hydrophobic
interior of the gel phase has a dielectric constant similar to that
of alcohols.19 With lower amounts of AcOH (e.g., 1% AcOH-
water), the primary aggregation concentration (CAC1 ≈ 25 µM)
was lower and I3/I1 values were higher (Table 2). Higher I3/I1
values indicate that the pyrene is in a more hydrophobic
environment (away from the aqueous phase) in both aggregates
and gels prepared in lower amounts of cosolvents in water. This
is an example of a more pronounced hydrophobic effect of water
when the concentration of organic cosolvent is lower. The
absence of the excimer band for pyrene in the gel phase (i.e.,
at a higher concentration of 1) could be explained as follows.
The excimer formation dynamics is governed by two factors
characterized by two rate constants: k1 (viscosity dependent)
and k2 (solvent polarity dependent).20 The first process (k1) of
bringing two pyrene (incorporated within the aggregates) closer
to form an encounter complex is diffusion controlled and would
be dependent on the solvent viscosity. If the viscosity is higher
2
mean-square distance (
x
∆x ) over which a quencher can dif-
fuse during the excited-state lifetime (τ) is given by
2
x
∆x ) (2Dτ)
(4)
(5)
x
k ∝ D ∝ 1/η
q
With lower amounts of acetic acid (1% v/v), due to the enhanced
hydrophobic effect (compared to that of 20% AcOH-water),
the aggregation is more pronounced (Table 1). Therefore, pyrene
3
is even more effectively protected from quenching by O2. A
relatively nonpolar microenvironment around pyrene is also
supported by a higher I3/I1 value (Table 2). In this case, the
gelling condition seems to be more important than the gelator
concentration. In the gel phase, the fluorescence lifetime of
pyrene (τ ≈ 324 ns) approaches the value observed in
deoxygenated cyclohexane (∼400 ns).
A partially water-soluble fluorescent probe of ANS was also
used to study the aggregation behavior of 1 since it has an
2
2
intriguing luminescence property. Fluorescence of ANS is
largely quenched in aqueous media due to the dynamic
collisional quenching, whereas in the presence of aggregates
(having hydrophobic cavities/clefts), fluorescence intensity
increases. Results obtained from ANS-binding studies (under
equilibrium conditions) also supported the secondary aggregation
model for the gelation of 1 (Figure 4). A small difference in
the critical aggregation concentrations determined by pyrene
and ANS can be attributed to the fact that ANS may have
additional electrostatic interactions with the aggregates, resulting
in a change in the emission property when bound to pre-micellar
aggregates. In the gel state, the fluorescence intensity of ANS
increased severalfold (30-fold for [1] ) 5.3 mM in 20% AcOH-
water). Figure 1B shows a bright luminescent gel photographed
in a dark room by illuminating it with a long-wavelength UV
lamp. A blue shift of the emission maxima from 520 to 470
nm in the course of the gelation further supports the change in
the local environment of ANS. Both the increase in the intensity
and the blue shift of ANS fluorescence were attributed to the
trapping of ANS in the hydrophobic pockets created in the gel
phase. This hypothesis was further supported by fluorescence
anisotropy data of ANS in the gel phase. Steady-state fluores-
(e.g., in gels), the approach of two probe molecules (prerequisite
to an excimer formation) gets hindered. The probability of the
encounter complexes (loosely bound) being converted to the
excimer (stronger complex) would be higher if the solvent
polarity is high (1% AcOH-water > 20% AcOH-water)
enough to necessitate the association of two hydrophobic probe
molecules in the micellar aggregates (Scheme 2).
Pyrene has an unusually long fluorescence lifetime (∼400
ns in deoxygenated cyclohexane) because the S1 f S0 transition
is partially forbidden.21 The fluorescence lifetime of pyrene (τ)
is shortened in nondeoxygenated polar solvents of low viscosity
because the two major rate processes depopulate the excited
state; (i) nonradiative decay (knr) is more efficient in polar
solvents, and (ii) freely diffusing oxygen (triplet) quenches
15910 J. AM. CHEM. SOC.
9
VOL. 126, NO. 48, 2004

Molecular Hydrogel from a Tripodal Cholamide
A R T I C L E S
cence anisotropy (rss) is an excellent indicator of flexibility of
the (supra) molecular structure to which the probe is bound.
This reports the dampening of the rotational motion of the
fluorescent dye in the time-scale of fluorescence.23 Variable
temperature fluorescence anisotropy of ANS was measured, and
the anisotropy was found to diminish in the temperature range
of 40-50 °C (Figure 6). Upon gel melting, the rotational
mobility of ANS molecules increases, resulting in the drop in
the anisotropy. The variable temperature fluorescence anisotropy
of ANS exhibits a more gradual decrease compared to the sharp
decrease in the molar ellipticity observed in the variable
temperature CD. This is not surprising since these two experi-
ments probe two fundamentally different properties, namely,
the supramolecular structure of the gel, which is related to the
amplification in the chirality, and the diffusional fluidity of the
microstructure of the gel, which is related to fluorescence
anisotropy of a bound dye. Also, the hysteresis, with melting
and freezing points being distinctly different, is typified by the
Figure 8. Time-resolved anisotropy decay profiles of ANS at two different
times during the gelation of 1 (5.3 mM) in 20% AcOH-water at 25 °C. (a)
2
At 4 min after mixing (pink): φb ) 3.7 ns, φf ) 0.2 ns, and ø ) 1.11. (b)
2
At 126 min after mixing (blue): φb ) 11.0 ns, φf ) 1.0 ns, and ø ) 1.10.
Solid black lines are the fits.
“
gel-to-sol” phase transition, as observed by us earlier for
24
organogels.
mixing). According to the Stokes-Einstein equation (eq 2), the
change in the rotational correlation time (φ) could be explained
in terms of the volume of the rotating unit (V) and the
microviscosity (η) felt by the probe. Thus, the increase in the
longer rotational correlation time (φb), from 3.7 to 12.1 ns, is
ascribed to an increase in both the aggregate size and the
microviscosity experienced by the ANS molecules bound to the
aggregates of 1. Our data on the time course of the average
translational diffusion of the aggregates (from dynamic light-
scattering experiments, Figure 7) also support the present
observation on the dynamics of ANS incorporated within the
aggregate. The increase in both aggregate size and viscosity
would cause the dampening of both translational and rotational
dynamics (eqs 1 and 2). The formation of an intertwined network
in the gel phase causes the dampening of the tumbling of bound
ANS molecules. Additionally, the time-course of the dynamics
of free aqueous ANS (characterized by a short rotational
correlation time, φf) seems to be even more interesting. After 4
min of the mixing, φf was ∼0.2 ns, which is very close to the
value (∼0.1 ns) obtained for ANS in an aqueous solution of
low viscosity (∼1 cP). With gelation time, φf becomes progres-
sively longer, which is indicative of incremental change in the
microviscosity of the aqueous phase, suggesting the microstruc-
tural change caused by the rigidification of the aqueous pool
encompassed by the networked nanofibers. We would like to
point out that the information about the partial immobilization
of solvent molecules (expressed in terms of local viscosity)
around the gel network could be experimentally probed by the
time-resolved fluorescence. This rigidification is likely to be
caused by the entanglement of the long nanofibers (Figure 1)
during the gelation process.
3
.4. Insights into the Gelation Process Using a Double-
Kinetics Approach. The time-resolved decay of the fluores-
cence anisotropy of probes carries information on the molecular
tumbling in the picosecond to nanosecond time regime. In the
steady-state fluorescence measurements, retrieval of information
is limited due to the fact that steady-state parameters depend
on several molecular properties. Time-resolved fluorescence
anisotropy measurements, using a picosecond laser, provide
insights into the local and global dynamics of complex molecular
2
5
systems. Since the decay kinetics can be resolved as a sum of
exponentials, it is possible to selectively probe different local
environments in a microheterogeneous system, such as a gel.
We have earlier shown that a hydrophobic dye, like DPH, has
a single population (evident by a single-exponential fluorescence
anisotropy decay kinetics) in the gel phase solubilized in the
hydrophobic pockets, whereas a partially water-soluble probe,
like ANS, distributes between the aqueous phase and the
network of the gel (double exponential).15 ANS has a faster
tumbling (single-exponential decay yielding φ ≈ 0.1 ns) in
isotropic aqueous solutions of low viscosity. In the presence of
smaller aggregates (pre-micellar, below CAC1) of 1, ANS shows
bi-exponential decay kinetics in the fluorescence anisotropy
decay. In the micellar aggregates (above CAC1, below MGC;
see also Figure 4), longer rotational correlation time (φb for
bound ANS) increases to about 5 ns, which is attributed to the
increase in the aggregate size. However, the free aqueous ANS
does not show any significant increase in the rotational
correlation time (φf e 0.2 ns) in aggregates or in aqueous
solutions. In the gel state (above MGC), both the rotational
correlation times of ANS (φb and φf) showed marked increase.
Therefore, we envisioned that the change in the rotational
dynamics of ANS, partitioned in the aqueous phase as a function
of time during the gelation (kinetic experiments), would carry
the information on the changing interfacial water structure
around the gel fiber network during the gelation. The time-
resolved anisotropy decay profiles of ANS fluorescence at
different times (Table 4) indicate that the dynamics of ANS is
progressively slowed during the gelation process. The decay
profiles indicate the increase in both long and short rotational
correlation times upon gelation (Figure 8, 4 and 126 min after
3.5. Microviscosity and Bulk Viscosity. In light of our time-
resolved fluorescence data, it is necessary to address the issue
of microviscosity and bulk viscosity of the gel. The gel derived
from 1 (5.3 mM in 20% AcOH-water) does not flow, which
means the bulk (macroscopic) viscosity of the gel is much higher
(indeterminable using classical techniques) compared to that of
3
glycerol (∼10 cP at 25 °C), which flows. Small molecule
fluorescent probes, such as ANS, are expected to show
considerably long rotational dynamics (>100 ns, calculated
2
6
using eq 2), which is not possible to be accessed by the time
J. AM. CHEM. SOC.
9
VOL. 126, NO. 48, 2004 15911

A R T I C L E S
Mukhopadhyay et al.
window set by the fluorescence lifetime (tens of nanoseconds);
as a result, fluorescence anisotropy would show an incomplete
depolarization in the time-scale of fluorescence. However, it is
interesting to learn that, despite our gels having much higher
bulk viscosity, the time-resolved fluorescence anisotropy exhibits
complete depolarization with time constants of 12.1 and 1.0
ns. Therefore, we conclude that the microscopic dynamics is
much faster compared to that we anticipate by visual observa-
tions. For an isotropic medium, such as glycerol, microviscosity
experienced by a fluorescent probe is comparable with the bulk
viscosity of the fluid. The enormous increase in the bulk
viscosity of the gel is thus attributed to the formation of an
entangled network comprising of nanofibers. The microviscosity
of the aqueous phase around the network of nanofibers is only
about 10-fold higher compared to bulk water and far less
compared to the bulk viscosity of the gel.
mixture and filtered through a sintered glass funnel, and the filtrate
was washed with water (40 mL), saturated NaHCO
finally with water (40 mL). The organic layer was dried over anhydrous
Na SO and evaporated to dryness. The crude product was purified by
column chromatography (25 cm × 2.6 cm silica column) using 2%
EtOH-CHCl to yield 3.6 g (62%) of a colorless product: mp ) 179-
81 °C. H NMR (CDCl , 300 MHz) δ: 0.753 (s, 9H), 0.835 (d, J )
3
(40 mL), and
2
4
3
1
1
6
6
3
.3 Hz, 9H), 0.954 (s, 9H), 1.09-2.16 (m), 2.53 (m, 3H), 3.25 (m,
H), 4.72 (m, 3H), 5.075 (s, 3H), 5.268 (s, 3H), 6.434 (t, 3H), 8.027
s, 3H), 8.111 (s, 3H), 8.169 (s, 3H). C NMR (CDCl , 75 MHz) δ:
3
13
(
12.14, 17.56, 22.31, 22.76, 25.54, 26.54, 27.18, 28.52, 31.30, 31.45,
33.41, 34.25, 34.48, 34.40, 35.00, 37.65, 40.75, 42.92, 45.00, 47.46,
-
1
54.61, 70.63, 73.70, 75.22, 77.20, 160.56, 173.92. IR (neat, in cm
)
+
ν: 1716 (s), 1643 (m). MALDI-TOF MS: [M + H] calcd for
C
87
H
132
4
N O
21, 1569.9; found, 1569.5. [R]24
Compound 1: Compound 5 (750 mg, 0.5 mmol) was dissolved in
a 2% KOH/MeOH solution (50 mL), and the mixture was stirred for
4 h at room temperature. Volatiles were removed under reduced
pressure, and the residue was dissolved in 25% EtOH/CHCl (10 mL),
D
+39° (c 1, EtOH).
1
4. Conclusion
3
filtered, and evaporated to dryness. The crude product was purified by
We have extensively studied gels obtained from tripodal
column chromatography on neutral alumina (22 × 1.6 cm) with a
cholamide 1 in aqueous media using various techniques involv-
ing cryo-TEM, CD, steady-state and time-resolved fluorescence,
and dynamic light-scattering. Our equilibrium studies shed light
into the complex nature of the gel network. A picture of the
gel could be constructed as follows. Molecules of 1 aggregate
in an asymmetric manner to yield an entangled network structure
of nanofibers having a relatively immobilized aqueous compart-
ment. The kinetic data on the gelation addressed the issue related
to progressive ordering of the solvent molecules around the
network of nanoscale fibers. Our findings concerning the
incremental change in both aggregate size and microviscosity
of the aqueous pool will be useful for other hydrogel systems.
It would be interesting to undertake further studies to understand
the structure and dynamics of solvent molecules entrapped in
the gel network.
5
3
-15% gradient of EtOH in CHCl to yield 355 mg (56%) of a white
1
solid: mp ) 192-194 °C. H NMR (DMSO-d , 300 MHz) δ: 0.558
6
(s, 9H), 0.792 (s, 9H), 0.904 (d, J ) 1.2 Hz, 9H), 3.05 (m, 9H), 3.60
(br s, 3H), 3.76 (br s, 3H), 4.00 (m, 3H), 4.09 (m, 3H), 4.39 (m, 3H),
1
7.70 (t, 3H). H NMR (CDCl
3
, 300 MHz) δ: 0.680 (s, 9H), 0.879 (s,
9
3
H), 1.054 (d, J ) 5.7 Hz, 9H), 1.22-1.55 (m, 18H), 1.75 (br m),
.75 (m, 3H), 3.82 (m, 3H), 3.96 (m, 3H), 6.91 (m, 3H). C NMR
1
3
6
(DMSO-d , 75 MHz) δ: 12.36, 17.13, 19.14, 22.61, 22.84, 26.22, 27.30,
2
4
8.56, 30.40, 30.48, 31.70, 32.66, 34.40, 34.90, 35.20, 36.93, 41.34,
1.53, 45.76, 46.29, 53.62, 62.79, 66.28, 67.75, 70.45, 71.04, 172.72.
-
1
IR (Nujol, in cm ) ν: 3338 (m br), 1642 (m), 1549 (w). MALDI-
TOF MS: [M + H] calcd for C78
+
132 4
H N O12, 1318.0; found, 1318.2.
HRMS: calcd1317.9920;found1317.9552.Anal.CalcdforC78
132 4 12
H N O ‚
5H
2
O: C, 66.54; H, 10.16; N, 3.98. Found: C, 66.72; H, 9.77; N, 3.75.
24
[R] +27° (c 2, EtOH).
D
Compound 8: Deoxycholic acid 6 (5.18 g, 13 mmol) was dissolved
in formic acid (15 mL, 0.39 mol), and the mixture was stirred at 55 °C
for 7 h. Excess formic acid was removed under reduced pressure. Then,
the free-flowing solid was crystallized from an ethanol-water mixture
(44/22 mL) and filtered through a Buchner funnel. Then, it was washed
5
. Experimental Section
5
.1. Syntheses and Characterizations. All of the reactions were
carried out in oven-dried round-bottom flasks. TLC was done using
precoated TLC plates (0.25 mm silica gel, with fluorescent indicator
UV254) obtained from Aldrich. After the elution, TLC plates were
developed using Libermann-Buchard reagent. For column chroma-
tography, commercial grade solvents were distilled prior to use. Gravity
columns were run with silica gel (100-200 mesh) and neutral alumina
with an ethanol-water mixture (2:1) and dried in a desiccator for 2
1
days to yield 5.24 g (88%): mp ) 189-190 °C. H NMR (CDCl
3
, 300
MHz) δ: 0.73 (s, 3H), 0.82 (d, J ) 6.0 Hz, 3H), 0.90 (s, 3H), 0.97-
2.36 (m), 4.81 (m, 1H), 5.23 (br s, 1H), 8.01 (s, 1H), 8.11 (s, 1H).
Compound 10: To dry CH Cl (10 mL), taken in an RB flask and
2
2
(
∼150 mesh, 58 Å) obtained from Aldrich.
cooled in an ice-water bath, were added diformyl deoxycholic acid
(5.01 g, 11 mmol), DCC (2.83 g, 13 mmol), and DMAP (1.5 g, 12
mmol), and the mixture was stirred. After 10 min, tris(2-aminoethyl)-
amine (325 µL, 2.2 mmol) was addded. The mixture was stirred at
Compound 7: Cholic acid 5 (5.21 g, 12.5 mmol) was dissolved in
formic acid (19 mL, 0.4 mol), and the mixture was stirred at 55 °C for
h. Excess formic acid was removed under reduced pressure, and the
free-flowing solid was crystallized from an ethanol-water mixture (50/
5
room temperature for 18 h. CHCl
mixture. The mixture was filtered through a sintered funnel, and the
filtrate was washed with water (40 mL), saturated aqueous NaHCO
3
(30 mL) was added to the reaction
7
1
5 mL). The crystals were washed with an ethanol-water mixture (1:
.5) and dried in a desiccator to yield 5.23 g (83%) of a white solid:
3
27
1
mp ) 203-204 °C (lit. mp 206-207 °C). H NMR (CDCl3, 300 MHz)
δ: 0.762 (s, 3H), 0.852 (d, J ) 6.6 Hz, 3H), 0.940 (s, 3H), 1.05-1.18
(40 mL), and finally with water (40 mL). The solvent was evaporated,
and the residue was dried in vacuo. The crude product was purified by
column chromatography (20 × 2.5 cm silica column) using 2% EtOH-
(
(
(
2
4
m, 3H), 1.27-2.43 (m), 4.68-4.73 (m, 1H), 5.071 (br s, 1H), 5.27
br s, 1H), 8.025 (s, 1H,), 8.108 (s, 1H,), 8.165 (s, 1H,). 13C NMR
CHCl to yield 1.60 g (50%) of a colorless product: mp ) 142-143
3
1
CDCl
3
, 75 MHz) δ: 12.13, 17.42, 22.31, 22.74, 25.51, 26.54, 27.12,
8.52, 30.37, 30.80, 31.30, 34.25, 34.41, 34.49, 34.68, 37.69, 40.76,
2.94, 45.00, 47.18, 70.66, 73.72, 75.25, 160.50, 160.59, 179.79.
Compound 9: To dry CH Cl (10 mL), taken in a 50 mL RB flask
and cooled in an ice bath, were added triformyl cholic acid 7 (6.7 g,
3.6 mmol), DCC (3.9 g, 18.9 mmol), and DMAP (1.6 g, 13.1 mmol),
3
°C. H NMR (CDCl , 300 MHz) δ: 0.746 (s, 9H), 0.835 (d, J ) 6.3
Hz, 9H), 0.928 (s, 9H), 1.26-1.90 (m), 2.53 (br t, 4H), 3.25 (br d, J
) 4.2 Hz, 6H), 4.80 (m, 3H), 5.25 (br s, 3H), 6.40 (br t, 3H), 8.026 (s,
3H), 8.131 (s, 3H). 13C NMR (CDCl
2
2
3
, 75 MHz) δ: 12.40, 17.56, 22.31,
22.76, 25.54, 26.54, 27.18, 28.52, 31.30, 31.45, 33.41, 34.25, 34.40,
34.48, 35.00, 37.65, 40.75, 42.92, 45.00, 47.46, 54.61, 70.63, 73.70,
1
-
1
and the mixture was stirred. After 10 min, tris(2-aminoethyl)amine (400
µL, 2.7 mmol) was added. The reaction was monitored by TLC (6%
EtOH-CHCl
75.22, 77.21, 160.56, 160.60, 173.92. IR (neat, in cm ) ν: 1721 (s),
+
1642 (m). MALDI-TOF MS: [M + H] calcd for C84
H N
132 4
O
15
,
(30 mL) was added to the reaction
1438.0; found, 1438.4. [R]24
+63° (c 1, EtOH).
D
3
). After 12 h, CHCl
3
15912 J. AM. CHEM. SOC.
9
VOL. 126, NO. 48, 2004

Molecular Hydrogel from a Tripodal Cholamide
A R T I C L E S
Compound 2: Compound 6 (550 mg, 0.4 mmol) was dissolved in
a 2% KOH-MeOH solution (50 mL), and the mixture was stirred at
room temperature for 14 h. The volatiles were pumped off, and the
The crude product was purified by column chromatography (22 ×
3
4.6 cm silica column) using 2% EtOH-CHCl to yield 0.5 g (59%)
1
3
of a colorless product: mp ) 101-103 °C. H NMR (CDCl , 300
residue was dissolved in 25% EtOH-CHCl
evaporated to dryness. The crude product obtained was purified by
3
(5 mL), filtered, and
MHz) δ: 0.745 (s, 3H), 0.837 (d, J ) 6.3 Hz, 3H), 0.925 (s, 3H),
1.0-2.1 (m), 2.236 (s, 6H), 2.408 (t, J ) 5.4 Hz, 2H), 3.315 (dd,
column chromatography on neutral alumina (22 × 1.6 cm) with 10%
J
1
) 5.4 Hz, J ) 5.7 Hz, 2H), 4.83 (m, 1H), 5.25 (br s, 1H), 6.01
2
1
3
EtOH-CHCl
3
to afford 284 mg (56%) of a colorless product: mp )
(br s), 8.028 (s, 1H), 8.131 (s, 1H). C NMR (CDCl , 75 MHz)
3
1
1
(
2
(
64-165 °C. H NMR (DMSO-d
s, 9H), 0.790 (d, J ) 5.7 Hz, 9H), 0.938-2.007 (m), 2.35 (br d, 4H),
.94 (m, 6H), 4.07 (m, 3H), 4.12 (m, 3H), 7.57 (t, 3H). 13C NMR
DMSO-d , 75 MHz) δ: 12.46, 17.10, 23.10, 23.58, 26.18, 27.05, 27.23,
6
, 300 MHz) δ: 0.458 (s, 9H), 0.718
δ: 12.24, 17.44, 22.80, 23.32, 25.63, 25.78, 26.35, 26.66, 27.24,
31.32, 31.98, 33.24, 33.90, 34.08, 34.56, 34.82, 35.48, 36.53,
41.62, 44.94, 47.36, 49.13, 57.80, 73.97, 75.91, 160.45, 160.53, 173.35.
IR (neat, in cm ) ν: 1724 (s), 1643 (m). MALDI-TOF MS: [M +
H] calcd for C H N O , 519.5; found, 519.8. [R]
-
1
6
+
24
2
8.64, 30.24, 31.72, 32.64, 32.95, 33.84, 35.17, 35.70, 36.31, 37.00,
+62° (c 1,
30
50
2
5
D
4
1.65, 46.00, 46.31, 47.45, 56.08, 70.00, 71.07, 79.19, 172.78. IR
EtOH).
-
1
(Nujol, in cm ) ν: 3365 (m), 1644 (m). MALDI-TOF MS: [M +
Compound 4: Compound 12 (150 mg, 0.2 mmol) was dissolved
in a 2% KOH-MeOH solution (2 mL), and the mixture was stirred
for 14 h. The volatiles were removed; the residue was dissolved in
+
Na] calcd for C78
H] calcd 1270.0072; found 1270.0249. Anal. Calcd for C78
H N
132 4
9
O , 1292.0; found, 1293.6. HRMS: [M +
+
H
132
N
4
9
O ‚
4
3
2
H O: C, 69.81; H, 10.51; N, 3.17. Found: C, 70.03; H, 10.36; N,
2
5% EtOH-CHCl
evaporated. The crude product was purified by column chromatography
on neutral alumina (18 × 2.6 cm) with 10% EtOH-CHCl as eluent
to yield 75 mg (60%) of a colorless product: mp ) 114-115 °C. H
NMR (CDCl , 300 MHz) δ: 0.677 (s, 3H), 0.907 (s, 3H), 0.970 (d, J
6.0 Hz, 3H), 1.0-1.8 (m), 2.236 (s, 6H), 2.414 (t, J ) 5.1 Hz, 2H),
3
(10 mL) and filtered, and the filtrate was
2
4
.42. [R]
Compound 11: To CH
550 mg, 1.1 mmol), DCC (290 mg, 1.5 mmol), and DMAP (141 mg,
.1 mmol) at 0 °C. N,N-Dimethylethylenediamine (110 µL, 1.0 mmol)
D
+52° (c 2, C
H
2 5
OH).
2
Cl (2 mL) were added triformylcholic acid
2
3
(
1
1
3
was added to the mixture after 5 min. The mixture was stirred at
room temperature for 18 h. The reaction mixture was diluted with
)
3
13
.33 (m, 2H), 3.60 (m, 3H), 3.97 (br s, 3H), 6.55 (br m). C NMR
2
0 mL of CH
water (20 mL), saturated aqueous NaHCO
20 mL). The organic layer was dried over anhydrous Na
2
Cl
2
. It was filtered, and the filtrate was washed with
(20 mL), and water
SO ; the
(CDCl , 75 MHz) δ: 12.69, 17.39, 23.10, 23.68, 26.13, 27.12, 27.48,
3
3
2
3
8.59, 30.40, 31.63, 33.31, 33.56, 34.09, 35.22, 35.29, 35.95, 36.42,
(
2
4
6.53, 42.03, 45.00, 46.45, 46.77, 48.14, 58.00, 71.53, 72.99, 174.99.
solvent was evaporated, and the residue was dried in vacuo. Purifica-
tion of the crude product by column chromatography (12 × 2.5 cm
-1
IR (Nujol, cm ) ν: 3328 (m), 1652 (m). MALDI-TOF MS: [M +
H] calcd for C28
found 463.3902. [R]
+
H
50
N
2
24
O
3
, 463.4; found, 463.8. HRMS: calcd 463.3899;
+35° (c 1, EtOH).
silica column) using 3% EtOH-CHCl
3
yielded 150 mg (45%) of a
D
1
colorless product: mp ) 143-144 °C. H NMR (CDCl
δ: 0.753 (s, 3H), 0.840 (d, J ) 6.3 Hz, 3H), 0.897 (s, 3H), 0.95-
.27 (m), 2.334 (s, 6H), 2.428 (t, J ) 6 Hz, 2H), 3.320 (dd, J ) 5.4
Hz, J ) 6.0 Hz, 2H), 4.67 (m, 1H), 5.06 (m, 1H), 5.16 (m, 1H), 6.35
br m, 1H), 8.026 (s, 1H), 8.105 (s, 1H), 8.162 (s, 1H). 13C NMR
CDCl , 75 MHz) δ: 12.14, 17.46, 17.56, 22.31, 22.76, 25.54, 26.54,
7.18, 28.52, 31.30, 31.45, 33.41, 34.25, 34.40, 34.48, 35.00, 37.65,
0.75, 42.92, 45.00, 47.46, 54.61, 70.63, 73.70, 75.22, 160.56, 173.92.
3
, 300 MHz)
5
.2. Test of Gelation and Measurements of Tgel. Gelation tests
were performed in 5 mm diameter Pyrex test tubes. The inverted tube
2
1
2
8
method was adopted to measure the Tgel of the samples. For all Tgel
measurements, 5 mm diameter sealed test tubes with 0.5 mL of solvents
were used.
2
(
(
3
5
.3. Cryo-TEM. We prepared cryogenic temperature transmission
2
4
electron microscopy (cryo-TEM) specimens in a controlled-environment
vitrification system (CEVS), as previously described.2 Specimens were
quenched from 25 °C and 100% relative humidity to prevent solvent
loss during specimen preparation. We imaged the vitrified specimens
in a Philips CM120 cryo-dedicated TEM, equipped with an Oxford
CT3500 cryo-holder system to maintain the specimens in the TEM at
about -180 °C. The TEM was operated with an acceleration voltage
of 120 kV. Images were recorded digitally with a Gatan MultiScan
9a
-
1
IR (neat, in cm ) ν: 1722 (s), 1643 (m). MALDI-TOF MS: [M +
H] calcd for C31
+
24
H
50
N
2
O
7
, 564.4; found, 564.6. [R]
D
+45° (c 1,
EtOH).
Compound 3: Compound 11 (150 mg, 0.2 mmol) was dissolved
in a 2% KOH-MeOH solution (2 mL), and the mixture was stirred
for 14 h. The volatiles were removed; the residue was dissolved in
2
5% EtOH-CHCl
evaporated. The crude product was purified by column chromatography
on neutral alumina (18 × 2.6 cm) with 10% EtOH-CHCl as eluent
to yield 75 mg (60%) of a colorless product: mp ) 159-162 °C. H
NMR (CDCl , 300 MHz) δ: 0.666 (s, 3H), 0.880 (s, 3H), 0.999 (d, J
5.4 Hz, 3H), 1.09-2.17 (m), 2.243 (s, 6H), 2.419 (t, J ) 6.0 Hz,
H), 3.310 (dd, J ) 6.0 Hz, J ) 6.9 Hz, 2H), 3.71 (m, 1H), 3.82 (m,
H), 3.94 (m, 1H), 6.61 (br t, 1H). C NMR (CDCl
3
(10 mL) and filtered, and the filtrate was
7
91 cool-CCD camera system.
5
.4. Circular Dichroism. Variable temperature CD spectra were
3
1
recorded using a JASCO J715 spectropolarimeter, equipped with a
Peltier temperature controller (JASCO PTC-348W1). All of the samples
were filtered through a 0.45 µm filter into a 0.5 mm path length quartz
cuvette and were allowed to form a gel. For variable temperature CD
experiments, samples were equilibrated for 10 min at every temperature
at which measurements were carried out.
3
)
2
1
1
3
5
1
2
1
3
3
, 75 MHz) δ:
2.46, 17.50, 22.50, 23.32, 26.34, 27.61, 30.53, 31.70, 33.10, 34.76,
4.80, 35.42, 35.50, 36.75, 39.49, 39.64, 41.61, 45.14, 46.36, 46.41,
5.5. Steady-State Fluorescence. Steady-state fluorescence spectra
were recorded on a Perkin-Elmer luminescence spectrometer (LS-50B).
ANS was obtained from Fluka (>98%) and was used without further
purification. ANS stock solution (1 mM) was prepared in double
distilled water, which was diluted to a final concentration of 10 µM
for all of the experiments. Samples were excited at 365 nm, and
emissions were measured at 475 nm. For fluorescence anisotropy (eq
-
1
8.11, 63.72, 68.38, 71.75, 73.02, 77.26, 174.38. IR (Nujol, in cm
)
+
ν: 3347 (m), 1650 (m). MALDI-TOF MS: [M + H] calcd for
28 50 4 2
C H O N , 479.4; found, 480.8. HRMS: calcd 479.3848; found
2
4
4
79.3821. [R]
Compound 12: To dry CH
deoxycholic acid (1.1 g, 2.4 mmol), DCC (0.6 g, 2.9 mmol), and DMAP
0.4 g, 3.2 mmol) under cold conditions. The mixture was placed in an
ice-water bath, and after 5 min, N,N-dimethylethylenediamine (220
µL, 2.0 mmol) was addded. The mixture was stirred for 16 h. CHCl
D
+32° (c 1, EtOH).
2
Cl
2
(2 mL) were added diformyl
(
⊥
6) measurements, perpendicular components of the emitted light (I )
were corrected with respect to the G factor of the instrument. A water
circulator (Julabo F25) connected to the sample accessory using
thermally insulating tubes was used for controlling the temperature.
For variable temperature anisotropy experiments, samples were equili-
brated for 10 min at every temperature at which measurements were
carried out.
3
(30 mL) was added, and the mixture was filtered. The filtrate was
washed with water (30 mL), saturated aqueous NaHCO
water (30 mL). The organic layer was dried over anhydrous Na
3
(30 mL), and
SO
The solvent was evaporated, and the residue was dried in vacuo.
2
4
.
J. AM. CHEM. SOC.
9
VOL. 126, NO. 48, 2004 15913

A R T I C L E S
Mukhopadhyay et al.
I| - GI⊥
I| + 2GI⊥
were excited at 307 nm, and emissions were measured at 490 nm. For
all of the measurements, 10K counts were collected. For time-resolved
anisotropy measurements, the emission data were collected at 0 and
r )
(6)
9
0° with respect to the excitation polarization, and the time-resolved
Pyrene was obtained from Aldrich and was purified on a silica
column using petroleum ether followed by a recrystallization from
methanol. A saturated solution of pyrene was prepared in water by
sonication for 2 h followed by filtration. The concentration of pyrene
in saturated solution was found to be 0.5 µM (UV method). Samples
anisotropy decays were analyzed by globally fitting I
|
(t) and I
⊥
(t):
I|(t) ) I(t)[1 + 2r(t)]/3
(7)
I⊥ (t) ) I(t)[1 - r(t)]/3
(8)
were excited at 337 nm. I
respectively. Excimer intensities were measured at 475 nm.
.6. Fluorescence Lifetime Measurements. Time-resolved fluo-
rescence decays were measured using a N flash lamp of short duration,
equipped with a time-correlated single photon counting (TCSPC)
obtained from IBH. Full width at half-maxima (fwhm) of the pulse
was ∼1.1 ns. All data were deconvoluted with respect to the lamp
profile and analyzed using IBH software that was provided with the
instrument. Data were fitted using single- or multiexponential fits using
the nonlinear least-squares method. Fluorescence lifetime of pyrene in
deoxygenated water was 174 ( 5 ns (reported previously as 175 ns).30
Pyrene solution was deoxygenated using three cycles of freeze-pump-
thaw. For micelle and gel samples, pyrene solution was used without
deoxygenation.
1
and I
3
were measured at 373 and 384 nm,
The perpendicular component of the fluorescence decay was corrected
for the G factor of the spectrometer. I(t) is the fluorescence intensity
collected at the magic angle (54.7°) at time t. The anisotropy decays
were analyzed according to the two-population model, wherein ANS
was assumed to be distributed between two phases (free and bound).
5
2
3
1
The time-dependent anisotropy r(t) is represented by eq 9.
r(t) ) f (t)r (t) + f (t)r (t)
(9)
f
f
b
b
where f
f b
(t) and f (t) are the fractions of the emitted photons corre-
sponding to the free and bound ANS at time t, respectively; r
f
(t) and
r
b
(t) are the anisotropy decay functions of the free and bound ANS,
respectively.
r (t) ) r exp(-t/φ )
(10)
(11)
f
0f
f
5
.7. Dynamic Light-Scattering Experiments. Dynamic light-
scattering experiments were performed on a DynaPro-MS800 instrument
Protein Solution Inc., VA), which monitors the scattered light at 90°.
rb(t) ) r0b exp(-t/φb)
where r0f and r0b are the initial anisotropies, and φ
f
b
and φ are the
(
rotational correlation times of ANS in two phases (free and bound,
respectively).
Acetic acid and water were filtered through 0.02 µm filters followed
by centrifugation at 12 000 rpm for 20 min before adding the mixture
to the gelator. “Regularization” software, provided with the instrument,
was used to analyze the data to obtain the hydrodynamic radius. A
solution of bovine serum albumin (3 nm) was used as a standard.
f
The intensity fraction, f (t), is given by
f (t) ) [x [R exp(-t/τ ) + R exp(-t/τ )]]/[x [R exp(-t/τ ) +
f
f
f1
f1
f2
f2
f
f1
f1
R exp(-t/τ )] + x [R exp(-t/τ ) + R exp(-t/τ )]] (12)
f2
f2
b
b1
b1
b2
b2
5
.8. Time-Resolved Fluorescence Anisotropy Measurements.
f b
where x and x are the fractions of free and bound probes, respectively.
Time-resolved fluorescence decay measurements of the samples were
made using a high repetition-rate picosecond laser (frequency-tripled
Ti-Sapphire laser) coupled to a time-correlated single photon counting
τf1 and τf2 are the fluorescence lifetimes of free ANS, and τb1 and τb2
are the fluorescence lifetimes of bound ANS. The free and bound ANS
had two fluorescence lifetimes.
(TCSPC) setup. A Ti-Sapphire femto/picosecond laser (Spectra
Physics, Mountain View, CA) pumped by an Nd:YLF laser (Millenia
X, Spectra Physics) was used. Pulses of 1 ps duration of 921 nm
radiation from the Ti-Sapphire laser were frequency tripled to 307
nm by using a frequency doubler/tripler (GWU, Spectra Physics).
Fluorescence decay curves were obtained by using the TCSPC setup,
coupled to a microchannel plate photomultiplier (Model 2809u,
Hamamatsu Corp.). The instrument response function (IRF) was
obtained at 307 nm using a dilute colloidal suspension of dried nondairy
coffee whitener. The width (fwhm) of the IRF was ∼40 ps. The samples
Acknowledgment. We thank the Jawaharlal Nehru Centre
for Advanced Scientific Research (Bangalore) for financial
support, and Professor N. Periasamy (TIFR, Mumbai) for
providing us the software for the fluorescence decay analysis.
We also thank Professor P. K. Das (IISc, Bangalore) for helping
us with fluorescence lifetime measurements.
JA046788T
(
31) (a) Ludescher, R. D.; Peting, L.; Hudson, S.; Hudson, B. Biophys. Chem.
1987, 28, 59. (b) Srivastava, A.; Krishnamoorthy, G. Arch. Biochem.
Biophys. 1997, 340, 159.
(
30) Vethamathu, M. S.; Almgren, M.; Mukhtar, E.; Bahadur, P. Langmuir 1992,
8
, 2396.
15914 J. AM. CHEM. SOC.
9
VOL. 126, NO. 48, 2004